ML030690642
| ML030690642 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse  |
| Issue date: | 09/23/2002 |
| From: | Tom Gurdziel - No Known Affiliation |
| To: | Grobe J Division of Reactor Safety III |
| References | |
| Download: ML030690642 (10) | |
Text
9 Twin Orchard Drive Oswego, NY 13126 September 23, 2002 Mr. John A. Grobe, Director Division of Reactor Safety US Nuclear Regulatory Commission 801 Warrenville Road Lisle, IL 60532-4351
Dear Mr. John A. Grobe:
I was very happy to read on your web page on September 18t", the slides presented at the meeting that day. Please extend to those responsible, my thanks.
Item 1 I thought the goal individual (human) error rate of.5 per 10,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />, given on slide 21 would be more typical of a machinery failure rate. In other words, I thought it was too ambitious. I went through the slides looking for errors to see if this important work presented to the public and the NRC would meet this goal. And I found it did, if they spent 100,000 or more person-hours on preparing the slides. I found 4 errors on slide 13 where 4 bullets were inconsistent with the upper one. On slide 611 noticed that when a change was made, not all the old words were removed. It says: "...has been incorporated into is a part...".
So I found 5 errors. What I found disturbing is that I expect that the slides were prepared at the highest level in the organization and I can't believe that they were not previewed.
This would mean that either people don't have sufficient observation skills, won't correct their own work, or are afraid to bring mistakes to the attention of management. At least, those are some reasons that come to my mind. And, it might be interesting to find out how many of the multitude of review organizations had even one person identify any error.
Item 2 There used to be an excellent source of practical power plant knowledge. It was, I believe, written by Stone and Webster for EPRI. I would recommend it as a reference for plant equipment root cause investigators and those site personnel writing Work Requests on plant equipment. I am attaching two memos that provide a little more information.
3OCTT 2J
Item 3 I am increasingly uneasy over the reluctance that I see, (from where I sit), to form credible Quality Assurance organization. Monitoring quality is not an equal substitute, and neither is assessing quality. Doesn't Appendix B to Part 50 apply to USNRC Region This is my seventh letter. It needs no reply.
Thank yqD.
1/2'\\
Copy: D. Lochbaum
INTERNAL CORRESPONDENCE POFtA 112-2 R OM2O 6".01-013 FROM T.
J.
Gurdsiel TO L.
D. Kassakatis DISTRICT Nuclear Division DATE February 19# 1992 FILECODE SM-SA92-0031 SUBJECT RCM Information for Electrical Controls Table 10.7 of EPRI Power Plant Electrical Reference Series, Volume 10, "itemizes some telltale signs that show components are failing and require some form of maintenance or replacement."
It may be useful to you for RCM work.
(It is attached.)
It may also be a useful reference for training instructors.
T.
J.,urdxiel Safety An ysis Support
/noe Attachment 001604ZZ xc:
T. G. Kulczycky R.
F. Kirchner IPE File G. J. Gresock R. G. Smith
- 1. Waldecker M
- 1T 1.7
POWER PLANT ELECTRICAL REFERENCE SERIES VOLUME 10 ELECTRICAL CONTROL AND INSTRUMENTATION ELECTRIC POWER RESt.ARCH INSTITUTE
ELECTRICAL CONTROL AND INSTRUMENTATION 10-59 a single breaker to protect equipment designed( t*o operate at two different current levels or two different supply voltages, such as 125 Vdc or 110/220 V, 50/60 Hz.
10.5 MAINTENANCE Maintenance of electrical controls is usually in the form of calibration, cleaning, or re-placement be.
fore failure. lDuipment loses calibration during use because of inherent drift of set points, tempera lure and prussuure fluctuations, and the intrusion of1 dust and dirt. Ilist, dirt, or aky otlher foreign partlichles can jallt delicate mechanicaiovettlents, cause arcing and burning, and can short circuit printed circuit boards and contacts. Many discrete components have a finite life due to wear, fatigue, corrosion, or breakdown from environmental ag.
ing. The timely replacement of these components Can prevent failures of control circuitry and avoid a nuisance tril) or shutdown.
CALIBRATION
(:alibrating of a.my elect rical or elect ronhlc Controls or instruments should he performed using the manufacturer's calibration procedures. Most elec t rnluic vot.llilnitllts have calibration circuits or self Cialill'alill I't ilnt ilatrS dlsignlitI ill*t) Illtlit. (alilb a.
tion requiring the use of another instrument must he performed, it is important that the calibrating itislit-mel.nt be within its tolerance of Calibration and that it be equal to or more accurate than the instrument to be calibrated.
CLEANING Many types of malfunctions can he caused by the intrusion of foreign particulates or the natural buildup of unwanted matter, such as carbon or by products of oxidation. Destruction of motor insu lation, nuisance tripping, sporadic operation, arc ing, burning, and noise are, to name a few, some Of tlif-ntts-tinllt MALl.'ltUltitn(1,s. W1i0tr4 1
.lOSSibhl, o011 tacts should be kept clean using a contact cleaning solution on a periodic basis. Cotton or fiber cloths should he used instead of abrasives where possible.
(Grit fromln sandpamlr and ittill e'y cloths and small metallic fil is trI'rullt stel WoolS Call aCCl*WiN'attE) cVl tact failure unless compressed air or contact clean.
ing fluid is used to completely clean the foreign matter from the surfaces of the contacts and the associated mechanical movements.
REPLACEMENT BEFORE FAILURE I
I "Many components, such as control switches, push buttons, circuit breakers, and fuses, can be diag nosed as a potential failure area before actual fail ure occurs. ibble 10-7 itemizes some telltale signs that show components are failing and require some form of nailintellalnc or Ire'ilactllinlt.
I II MISCELLANEOUS perioxlic inspretions and survteillance of (1ont1r(1ls Slid illstrunlen(*ts shoutld be niact' t0 mon0111iltor.lty degradation of equipment. These inspections should be made as often or more frequently in re.
mote areas or areas in harsh environments than inspections within the main control room. The con trol room instrumentation and controls are used and viewed daily; however, remote equipment, es pecially in outdoor locations, can degrade rapidly and fail if not properly attended. Rubber boots in stalled over push buttons or selector switches for Table 10.7 Troubleshooting Chart for Electrical Controls component
- Sympliom, Probable Cause '
PuSh buttons and Loose Wtwn parts Coftlol Switches Fuses Circuiat breakers Relays or contactors Indicating lights Sticky Corroded Part of foreign matter Spodc operation Con surace coroded or charred. contacts bent or loose Hot, charred Operating at or close to appearance rated interrupting capacity-fuse element weakened by thermal stress Hot, sporadic trips. Operating at or close to arcing rated interrupting capacity-fuse element veakened by thermal stress Mechanical elements loose or weakened; dust.
dirt or foreign mailer on contacts or ill muchum.ical movements; loose terlina tion 0o"n Hum or chatter.
Oust, dirt, or foreign matter arcing on contacts or in nwctanican movoments, loose to*irinaion 10ri Fickserig Loose bulb, bad socket.
loose termination joint
INTERNAL CORRESPONDENCE' MVNI&R FOIM 11324f0240 541-41 NU FROM DISTRICT Nuclear Division TO Distribution OATE October 2, 1989 FILE CODE SUBJECT Electrical Noise Reduction Reference Today I discovered Electrical Power Research Institute publication EL-5036, Volume 14, Electrical Noise Reduction.
The site library will be obtaining a copy.
Until then you can borrow the copy of Mike Cowden in Salina Meadows at Extension 7307.
1 am attaching both an Abstract of its contents and two pages which contain troubleshooting flowcharts as examples of its usefulness.
T. J.Gurdel Associate Senior Nuclear Engineer TJG/cmd 5352W Attachment Distribution D. Greene T. Kulczycky P. Mazzaferro M. Goldych R. Jenkins P. Carroll, Security C. Fischer R. Morey T. McMahon R. Pasternak A. Rivers C. Beckham K. Ward D. F. Helms L. Lagoe K. Sweet S. Doty J. Kinsley M. Cowden G. tiltse xc:
Records Management
POWER PLANT ELECTRICAL REFERENCE SERIES VOLUME 14 ELECTRICAL NOISE REDUCTION ELECTRIC POWER RESEARCH INSIIIUIE
ABSTRACT Power plants by nature are electrically noby en vironments for systems using low-level signals. All power lines, motors, generators, relays, and other current-handling devices radiate magnetic fields of erratic and varying strengths. Common grounding and return conductors create resistive or conduc tive paths in signal, control, and power circuits.
Radio-frequency disturbances occur when fixed and portable communication radios are operated.
Electric power system can create electromagnetic field disturbances when their operation produces arcs, as during the energization or Interruption of circuits or the switching of shunt capacitors.
All of these disturbances create interference that ranges from a degradation of the signal-to-noise ratio on a circuit to complete disruption or loss of service for critical plant systems These systems include:
- Communications systems
- Digital computers and data acquisition systems
"* Protective relays
"* Instrumentation and controls
"* Nuclear instrumentation and reactor protec tion systems This volume categorizes electrical noise sources by type, level, and coupling mechanism. It describes the coupling mechanisms (electric field, magnetic field, and resistance) and presents methods for predicting the level of interference for mamy caes of woupbln& ncludin radiated and conducted cases.
The purpose of this volune is to aid the plant engineer in selecting the most effective means to prevent noise from disrupting sensitive circuts.
A troubleshooting guide is included to help pin point noise problems in existing installations and help mitigate electrical noise.
ELECTRICAL NOISE REDUCTION 1445 t.U3 k4atfuncnc.n 1iubIeshootzflO Chart
-. u-------
Iskep the amount of exposed wire at the sampling process. Any high-frequency no ound p
end to a minimum: (Get as close as beyond the filter would therefore be erated in possib the measurement point.)
ternally. Measurement will start output and SShortthe s robes together to measure work backwards towards input. Using the the inherentno f te s
t is previously shown oscil arrangement (Fig much less than the n to be measured.
ure 14-40), p it the following steps:
Tlke measurements at all p
and outputs
- 1.
scope across the input to the A/D including power supply b necIf in order p
rter. It is necessary to trigger the scope to isolate the location of the noie int
.sys-from the -dgtie command.
tematically disconnect leads and replace
- 2. Set the sweep time to coincide with the A/D sourcthe of raunise eing on d
ucto conversion time (usually 10 is/dlivision).
source o( nob being conducted into e[l
- t.
rs If noise is found on a circuit, check each end di scope plays the noise that the A/D con to determine the characteristics he noise sig-any noise present other than nal. Record the value of noise nt, and use a that at digitizing time will not be seen.
scope camera to photogf signal. This is use-With constant any variation of ful for the evaluation noise source.
sinlmost likely be caused by circuitry ahead o A/D converter.
MEASUREMEN PPROACH FOR A TYPICAL
- 4. If noise is nd connect scope across the input of the commutator p.e-and-hold The fo ng example demomstrates the approach amplifier). Thgger the scope f the corn to snuring noise for a typical system. The sys-mutator drive pulse.
chosen (Figure 14-41) (29) is a data acquisition system. Most data acquisition systems provide low-
- a. Set the sweep to coincide with the sa le pass filters (approximately 500 Hz) prior to the time (typically 5 las/division).
14-46 POWER PLANT ELECTRICAL REFERENCE SERIES y
Da mage dt SAnalyze daomage to determine s 11 4
Power Input f eed line (Block)
Power (White)
Ground (Greenl I
Signa wire' FO
. 14M4 NISe Mea Configuratb
- 6. Record the value noise using a scope camera.
n
- 7. Disable mutator operation and place a
- mp the commutator input to output.
S.
rd the noise. A difference between this value and the previous value would indicate Iource OuptI"inr e
breokdown protection to service that the noise is generated by the commu tor. Possible causes could be malfuncti ng commutator switches, improper s I and shield termination, and power s ply noise pickup. Check these.
- 9. If the noise Is at the in the commuta tor, move the scope to the output of the filter. igger scope internally. There should be no beyond the bandpass of the filter.
- 10. From point on, the approach is to leave t
pe connected to the rdter output, a connection, and apply a short across the line. Using this technique, work toward transducer. An increase in noise at a par point indicates that the noise genera tor ted between this point and the previous int.
DETECTION OF UND LOOPS ON AN INSTRUMENTATION S EM Ground loops can be cau impr(p tion of shields, signal wires, a
The detection of ground loops al pie. The noise signal will be moostly 7 )
tive signal but may also contain a harmonic, as shown in Figure 1442. If 6' 1r connec gro*unds.
ways sim Hz repeti third I
I